The initially AMD Radeon RX 6000 sequence graphics playing cards have been unveiled just a several several hours ago and they look gorgeous offered their technical specs, performance numbers and especially, the selling prices. But the announcement was lacking a key factor which we had been all hoping to get a lot more specifics which is the ray-tracing effectiveness of the upcoming-generation RDNA 2 lineup.
AMD’s Radeon RX 6000 “RDNA 2” Large Navi GPUs Ray Tracing Overall performance Explored – 1st Gen AMD Ray Accelerator Cores vs 2nd Gen NVIDIA RT Cores
While AMD failed to provide any specific numbers related to ray tracing efficiency, a general performance measurement on 1 of the three RDNA 2 “Huge Navi” GPU centered Radeon RX 6000 series graphics card has slipped by inside of the announcement slide deck. The quantities mention an AMD RDNA 2 graphics card which was tested inside the DXR Procedural Geometry Sample software with both of those hardware acceleration enabled and disabled.
Calculated by AMD engineering labs 8/seventeen/2020 on an AMD RDNA two centered graphics card, using the Procedural Geometry sample software from Microsoft’s DXR SDK, the AMD RDNA 2 centered graphics card will get up to 13.8x speedup (471 FPS) working with HW dependent raytracing vs working with the Application DXR fallback layer (34 FPS) at the same clocks. Overall performance may perhaps fluctuate.
New to the AMD RDNA 2 compute device is the implementation of a substantial-effectiveness ray tracing acceleration architecture regarded as the Ray Accelerator.
The Ray Accelerator is specialised hardware that handles the intersection of rays giving an get of magnitude improve in intersection overall performance as opposed to a software implementation.
With hardware acceleration disabled and just relying on the Software DXR fallback layer, the AMD Radeon RX 6000 graphics card scored 34 FPS. With hardware acceleration enabled & creating total use of the recently added Ray Accelerator cores, the AMD Radeon RX 6000 series graphics card observed a 13.8x speedup. The similar RDNA 2 graphics card scored 471 FPS in the software which is a large uplift in efficiency.
Evaluating to NVIDIA’s initial generation ray-tracing implementation, the GeForce RTX 2080 scores about 308 FPS. This means that the AMD RDNA 2 based Radeon RX 6000 sequence graphics card scores a 50% lead above NVIDIA’s to start with generation RTX graphics card. The GeForce RTX 2080 Ti scores close to 390 FPS which places the AMD Radeon RX 6000 graphics card twenty% ahead.
Comparing it to the Ampere dependent GeForce RTX 3080 is a whole unique story. Ampere introduced the next era of RayTracing cores which offer a enormous leap in efficiency as opposed to the initially era RTX twenty series playing cards.
NVIDIA GeForce RTX 3080 DXR Ray Tracing Overall performance:
A consumer on AMD’s subreddit posted the score of his stock ASUS GeForce RTX 3080 TUF Gaming graphics card which sent 630 FPS in the same benchmark. This shows the 2nd Era RT cores at about 33% a lot quicker than AMD’s 1st Gen Ray Accelerator cores. My own GeForce RTX 3090 scores all over 750 FPS which is all-around 50% speedier than what AMD’s Radeon RX 6000 graphics card scored in the exact same benchmark.
NVIDIA GeForce RTX 3090 DXR Ray Tracing General performance:
We continue to you should not know which graphics card was utilized for this comparison as the RDNA 2 lineup contains the 60 RA RX 6800, 72 RA RX 6800 XT and the 80 RA 6900 XT graphics card. As for the benchmark itself, it is most possible that the default settings and resolution have been utilised for this examination considering the fact that that was the case with my personal benchmark run.
You will find a good deal that needs to be reviewed about AMD’s ray tracing implementation and we hope AMD supplies us a lot more facts in the long run prior to the embargo lift which really should be someday in the coming thirty day period.
The publish AMD Radeon RX 6000 “RDNA 2 Significant Navi” GPU Ray Tracing Overall performance Specific – NVIDIA’s RTX 3080 With RT Cores 33% A lot quicker Than AMD’s Ray Accelerator Cores by Hassan Mujtaba appeared initial on Wccftech.